Content-based image retrieval using salient points

Salient point is an important point feature in the content-based image retrieval. A novel image retrieval based on salient point is proposed in this paper. Firstly, an improved algorithm is presented based on SPARSE(Salient Points Auto-Reduction using Segmentation), which using dynamic threshold segmentation algorithm in the between-class variance and within-class variance of the image segmentation. And then three color features and three texture features are adopted by image retrieval. Experimental results show that the improved method has better performance.